As customer expectations are rising and fraud risks are escalating, companies are reevaluating how they engage with consumers—particularly over the phone. While voice has long been viewed as an expensive and operationally complex channel, recent advances in AI are changing that perception. At the forefront of this transformation is Regal, a company founded in 2020 to modernize the voice experience through intelligent, enterprise-grade AI agents.
To better understand where the industry is headed, I spoke with Alex Levin, co-founder and CEO of Regal, about how his company is helping enterprises unlock scalable, automated voice interactions that not only reduce fraud but also enhance customer satisfaction. In our conversation, Levin shared insights into Regal’s technology, real-world deployments, and what it will take to automate over 90% of contact center interactions within the next decade.
Tell me about Regal, how long have you been around and what do you do?
My co-founder, Rebecca, and I were previously at a company with a large contact center, and we saw firsthand how much customers wanted to call us, but from our side, voice was a costly and difficult channel to operate. We tried all the tools out there, but couldn’t find one that really worked.
So, we left and started Regal in 2020 to bet on voice as the channel of the future and to improve the customer experience. That means shorter wait times, better resolution, higher NPS, and so on. We’re not just building AI agents. We’ve built an entire platform to orchestrate and manage voice calls, run A/B tests, and post-call analysis.
At its core, Regal builds voice AI agents for enterprises. We help companies create agents that are deeply customized for their customers and use cases.
How does Regal train its voice AI agents to balance fraud detection with maintaining a positive and natural customer experience—especially in high-emotion interactions?
I think there’s a big misconception around AI agents. If you pulled a random person off the street and put them on the phone without any training, they’d be terrible. Same with AI. But if you give AI the right training, guardrails, and input, it can actually perform better than a human.
For example, we work in highly regulated industries like collections and healthcare. There are strict rules around what can and can’t be done, and we can program those directly into the AI. We also operate in industries where fraud is common, and we’re able to analyze transcripts in real time to identify and flag them.
We even give our agents distinct “personalities” and tones. If you check out regal.ai/dogs, you’ll see how we’ve created different AI voices inspired by dog breeds to show how varied and expressive AI agents can be.
You mentioned over 90% of contact center interactions may be automated within a decade. What needs to happen—in terms of technology, regulation, and consumer trust—for that projection to become reality?
The biggest shift has already happened over the last 12 to 18 months: AI agents are now not only cheaper than human agents, they’re also just as good, or better.
What still needs to change? First, perception. Most people still associate voice automation with bad IVRs or outdated NLP systems. This new generation is completely different, and we work hard to demo that so people can experience the improvement firsthand.
Technologically, companies need to start building APIs instead of buttons for human agents. The AI needs to be able to take actions, such as update a payment, change an address, etc., and that requires an API.
From a regulatory perspective, this current administration has been pretty hands-off, so that’s not a blocker. And on the consumer side, even though the AI might not always sound perfectly human, it performs as well, or better, because it’s always available, highly capable, and always polite to the customer.
How does Regal’s approach to fraud detection differ from legacy systems, and what kind of real-time data signals or behavioral cues do your agents rely on to flag risks?
We work alongside many legacy systems, but it’s not the primary thing we’re responsible for. What we do differently is bring real-time transcription into the process. We can monitor what’s being said as it’s happening and flag the risk on the spot. The data lives in very sophisticated systems that can help identify the risk as well.
Why do customers choose your company?
Regal is best at one thing: voice AI agents, particularly for the enterprise. We’re focused on making the experience feel human. So, that means great audio quality, responsive agents that can take action, and integrations that just work.
We offer a self-serve product, but we also give customers significant hands-on support during implementation, free of charge. Most customers can get 80% of the way there on their own, but we’re there to help with the last 20%.
Our pricing is also attractive: we charge per minute, and we’re about 10% of the cost of a human agent in the U.S.
Beyond the agents themselves, we’ve built a powerful platform with deep capabilities: inbound/outbound orchestration, A/B testing, and post-call analysis. These are tools you won’t find on other platforms. They allow companies to run more sophisticated AI agent programs with Regal than with any other system.
In what ways do you see inflation and economic pressure accelerating the adoption of AI voice agents—not just to reduce fraud but also to help businesses maintain service levels with tighter budgets?
Put simply, contact centers have always been hesitant to offer voice support because it’s expensive. For the first time, AI makes voice affordable and high-quality, so I think that’s the primary motivation. AI is a deflationary force: it brings costs down. That frees up budget to invest in other areas like marketing, product development, and even reducing prices.
Meanwhile, human labor costs keep rising with inflation. On top of that, about 40% of contact center staff churn annually because most people don’t want to do the job. So, it’s not about firing human agents; it’s about not needing to replace them.
Can you walk us through a real-world deployment where Regal’s AI voice agent directly prevented fraud or delivered a measurable improvement in customer experience or efficiency?
A great example is our work in Medicare Advantage, a highly regulated industry with significant fraud risk. There are a lot of people trying to get around the system, so we help companies replace the initial screening process with AI agents. That allows legitimate customers to get through faster, with shorter wait times and a better overall experience.
And when someone is trying to game the system, our agents can automatically detect that in real time within the system, something that’s harder to rely on humans to catch consistently.
Conclusion
Regal’s vision reflects a broader industry shift: AI isn’t just automating contact centers—it’s redefining what customers expect from voice support. By investing in fraud detection, post-call analytics, and customizable agent personalities, Regal is helping companies deliver experiences that are not just efficient, but personalized and trustworthy.
As businesses contend with rising costs, labor shortages, and pressure to differentiate on service, Levin’s insights point to a future where voice is no longer a cost center—but a competitive advantage.
Learn how AI Agents can supercharge your company’s profits and productivity at TMC’s AI Agent Event in Sept 29-30, 2025 in DC.
Rich Tehrani serves as CEO of TMC and chairman of ITEXPO #TECHSUPERSHOW Feb 10-12, 2026 and is CEO of RT Advisors and is a Registered Representative (investment banker) with and offering securities through Four Points Capital Partners LLC (Four Points) (Member FINRA/SIPC). He handles capital/debt raises as well as M&A. RT Advisors is not owned by Four Points.
The above is not an endorsement or recommendation to buy/sell any security or sector mentioned. No companies mentioned above are current or past clients of RT Advisors.
The views and opinions expressed above are those of the participants. While believed to be reliable, the information has not been independently verified for accuracy. Any broad, general statements made herein are provided for context only and should not be construed as exhaustive or universally applicable.
Portions of this article may have been developed with the assistance of artificial intelligence, which may have contributed to ideation, content generation, factual review, or editing.
